Guidance on the conduct of pre-procedural checks of pregnancy status in UK women of child-bearing age has been developed to avert the risks of surgery and anaesthesia in a previously unidentified pregnancy. It was initially recommended that formal pregnancy testing should be selective in its application, depending on the risks posed by procedures. In response to 42 incidents, when pregnancy status was not established, the National Patient Safety Agency (NPSA) recommended: (a) tightening procedures to ensure that audited checking and recording of pregnancy status occurred; (b) consensual blanket testing of all menstruating women for pregnancy prior to surgical procedures.1 The NPSA acknowledged that ‘particular consideration applied to under 16-year-olds’, necessarily so since the mean age of menarche in the UK is 12.7 years, and UK teenage pregnancy rates remain high. Yet, as Donaldson et al report,2 knowledge and implementation of this guidance is incomplete and patchy. Consensual blanket testing of all under 16-year-olds prior to relevant procedures is a possible solution, but there are concerns as to whether this is a proportionate response, and how ethical and procedural issues should be addressed.

Relatively few UK under 16-year-olds become pregnant, with 70%–75% pregnancies occurring in over 15-year-olds. This suggests that an age-targeted selective approach to testing may be appropriate. The risk to pregnancy posed by surgery, anaesthesia or ionising radiation in UK teenagers has not been defined and has to be seen in the context of higher incidence of pregnancy loss and other adverse outcomes in this age group. Rates of foetal loss associated with surgery and anaesthesia in over 20-year-olds are similar to the rates of spontaneous pregnancy loss in the under 20 age group (approximately 7–11% in each case). No UK studies have defined the prevalence of positive pregnancy testing in teenage girls undergoing surgery. In the US pregnancy rates of between 0% and 1.2% have been recorded by preoperative consensual pregnancy testing in teenagers and young adults, but very few identified pregnancies were in those aged 15 or younger.3 These factors may have contributed to a move away from recommendations for routine testing by American anaesthetists.4

Pregnancy testing also has cost implications which may be emotional and psychosocial as well as financial. In the US, the cost of detecting a pregnancy by pre-procedural testing is approximately $3000 per case identified. The cost of an individual urine test in the UK is under £2, but in the absence of statistics on the total number of under-16s undergoing relevant procedures, neither the total cost nor the cost-per-case identified can be computed. It has been argued that, if the evidence for risk of harm from surgery or radiation in this age group is small, and the emotional cost of testing for children and families is high, the performance of tests, whose outcomes may have limited implications for the procedures to be undertaken, is hard to justify. In contrast, there may be considerable emotional costs occasioned by the absence of pregnancy testing if a miscarriage occurs, or if the foetus is exposed to the avoidable harms of surgery, anaesthesia or irradiation. Litigation claims from adults have arisen because of failure to perform pregnancy checks before surgical procedures but the foetus has very limited legal rights or recognition on which such claims might be made.

The sensitivity, specificity and timing of urine tests for pregnancy and the wider implications of possibly spurious results need to be carefully considered. Regardless of uncertain outcomes of early pregnancies, it is important that they are detected and thus provide patients and professionals with information on which to base choices, for example, whether to go ahead with a procedure or to delay it.

One argument advanced in support of blanket testing is the perceived unreliability of the history given by younger people, both in relation to their sexual activity or potential pregnancy status. Verbal recall of the last menstrual period may be inaccurate and unreliable in the presence of an irregular menstrual cycle. Reports of unreliability need to be set in context of what is known of sexual behaviour in adolescents, for example, age of first sexual intercourse, and the difficulty that some professionals have in addressing these sensitive questions (and which the current study reveals2). Matters may be further complicated –and frank discussion inhibited– by the presence of parents.

Even if patients are pre-warned that testing is routinely offered consent can neither be implied nor presumed. Nor can consent for testing, even with an opt-out provision, be regarded as a valid part of general surgical consent. Valid consent is necessary and must be adequately informed and freely given by a person who is competent to do so. Although children over the age of 12 year are presumed competent under Scottish law, in England and Wales the standard of Gillick competence applies to under-16s.5 The child must be able to fully understand the nature and purpose of what is involved, its risks and the consequences of alternatives for themselves and their family. It is not clear what action should follow refusal of testing by an adolescent (whether competent or not), how such a refusal should be interpreted and whether it can or should be overridden by parents.

Preoperative pregnancy testing also raises sensitive questions concerning an adolescent's sexual activity and her right to confidentiality. Professionals have a moral, if not common law obligation to warn children of the risks of sexual activity, and should encourage them to discuss matters with their parents or offer to do so on their behalf 5 but these suggestions may be rejected. The child's confidentiality should generally be respected, provided it is in their best interests to do so, and the professional believes that there is not a serious enough risk to their health, safety or welfare to warrant disclosure.

Disclosure of sexual activity, especially when accompanied by a positive pregnancy test, raises concerns that might otherwise have remained covert. In England and Wales children under 13 years cannot legally consent to sexual activity,6 and all disclosures/pregnancies in this age group are considered serious matters. Safeguarding concerns arise in older children (up to 18 years of age) where there are indications of non-consensual sexual activity or of violence or abuse. Even when such concerns do not arise, there may still be ethical dilemmas for professionals and families concerning management of risk-taking behaviours and family relationships that result from discussion/disclosure.

There is a clear need for national guidance on the checking of pregnancy status in UK girls undergoing investigations or procedures that pose risk to an undetected pregnancy. However, any recommendation on the use of blanket pregnancy testing post-menarche, even if it is intended to minimise these risks, should be based on quantitative and qualitative analysis of the harms and benefits in this particular population, its cost-effectiveness, considerations of fairness and proportionality and the ethical and legal consequences of implementation. Until this has occurred, it would seem appropriate that guidance should recommend ensuring that a practice involving sensitive questioning and selective testing should be properly implemented, recorded and audited, as remains the case for adults. Any future procedural guidance or policy on checking pregnancy status in the under-16s should be evidence-based, transparent, proportionate, fair and sensitive.